Mutations caused by DNA damage in the cell is a major cause of cancer. Familial mutations in colorectal cancer (CRC) account for less than 3% of all CRC cases, while other key tumor suppressors or proto-oncogenes remain unidentified. For sporadic CRC, pinpointing the causative mutation(s) within a cancer containing many mutations and epigenetic modifications is also incredibly difficult. Thus, it is likely that many genetic modifiers and causative mutations remain unidentified. Recent developments of mutagenesis strategies in the mouse now make it possible to rapidly identify causative mutations and modifiers involved in carcinogenesis. The objectives of this proposal are (1) to develop a transgenic mouse model of colorectal cancer through random insertional mutagenesis in the intestinal epithelium, and (2) to identify novel cancer modifier genes within a mouse model of CRC. Genes to be identified include novel cancer genes and pathways, modifiers of APC/Wnt signaling, and synergistic or cooperative mutations. This proposal details a novel forward genetics approach using a transgenic L1 retrotransposon mutagenesis system. Two mutagenic L1 cassettes will be used in this proposal: one containing a small bidirectional splice acceptor gene trap for interruption of gene expression (LOF module), and one containing a constitutive promoter and splice donor for over-expression of possible oncogenes (GOF module). Both transgene cassettes will employ the mouse Vil1 (Villin) promoter. This promoter is unique in its ability to drive specific high-level copy-number dependent expression throughout the entire intestinal epithelium: in stem cells, progenitors, and differentiated cells. Polyps from a transgenic mouse can be harvested and mutations readily identified by ligation-mediated PCR of genomic DNA or by 3' RACE of RNA. Relevance: This proposal describes a method employing powerful genetic tools developed in the mouse for specifically targeting the intestinal epithelium, the tissue where colorectal cancer originates. This approach can potentially identify hundreds of cancer gene mutations in a single mouse, as opposed to traditional methods where only one gene is mutated in a single mouse. Thus, one can rapidly identify many mutations that have the ability to cause colorectal cancer. This may uncover new mechanisms for how cancer initiates. [unreadable] [unreadable] [unreadable]